This is cross-posted from the GCRI website.
GCRI is currently welcoming inquiries from people who are interested in seeking our advice and/or collaborating with us. Inquiries can concern any aspect of global catastrophic risk. We welcome inquiries from people at any career point, including students, any academic or professional background, and any place in the world. People from underrepresented groups are especially encouraged to reach out.
We are especially interested in inquiries from people whose interests overlap with ours. For details of our interests, please see our publications, topics, and our current funded AI policy projects. We are most interested in fielding inquiries on the select projects outlined below. That said, we want to stress that this is an open call, and we encourage people to reach out to us if they are interested in any aspect of global catastrophic risk.
We also welcome inquiries from both colleagues we already know and people we have not met before. This open call is a chance for us to catch up with the people we already know as well as a chance to start a new relationship with the people we have not met before. It is also chance for anyone to talk with us about how to advance their career in global catastrophic risk, to explore potential synergies with our work, and to expand their networks in the global catastrophic risk community.
Participation does not necessarily entail any significant time commitment. It can consist of anything from a short email exchange to more extensive project work. In some cases, people may be able to get more involved by contributing to ongoing dialog, collaborating on research and outreach activities, and co-authoring publications. Some funding is available for people who collaborate with us on project work. We are most likely to fund work on the projects listed below, though there is no strict restriction on the scope of the projects we can fund. Details are available upon request.
This is the third time that GCRI has hosted an advising and collaboration program. Prior programs are documented here.
Individuals interested in speaking with us or collaborating with us should email Ms. McKenna Fitzgerald, mckenna [at] gcrinstitute.org. Please include a short description of your background and interests, what you hope to get out of their interaction with GCRI, a resume/CV or a link to your professional website, and where you are based. It would also be helpful to include any ideas for how you could contribute to GCRI’s projects.
The projects:
AI policy
We are doing a mix of research and outreach on AI policy. The research studies what policies could affect global catastrophic risk, including by affecting outcomes related to long-term AI. The outreach seeks to influence ongoing AI policymaking in the US and internationally. This work is being done in close collaboration with colleagues at other organizations interested in AI and global catastrophic risk. We welcome inquiries from people interested in these topics, especially those not already coordinating with others in the global catastrophic risk community on how to approach these issues.
Expert judgment on long-term AI
We are applying best practices in expert elicitation methodology to the study of long-term AI. Specifically, we are evaluating prior use of expert judgment on long-term AI and developing recommendations for future research. We welcome inquiries from people with background in expert elicitation methodology and people who are involved in research using expert judgment on long-term AI.
Forecasting global catastrophic risk
We are studying potential ways to forecast global catastrophic risk accurately. We are particularly interested in extending and adapting the methods associated with the Good Judgement Project to forecasting rare and unprecedented events like global catastrophes. We welcome inquiries from anyone interested forecasting global catastrophic risk.
Improving China-West relations
We are at the early stages of studying China-West relations as they relate to global catastrophic risk. Specifically, we are studying whether there are opportunities to improve relations that could significantly reduce global catastrophic risk. Our work is largely motivated by concerns about the potential danger of competition to develop advanced AI, but we recognize that improved relations could mitigate a range of global catastrophic risks. We are particularly interested connecting with people who have a background in China-West relations and people from China with a perspective on these issues.
National security dimensions of AI
We are doing work on the relationship between AI and nuclear weapons and on other military applications of AI, including autonomous weapons. We are particularly interested in connecting with people with up-to-date knowledge of the academic literature in and the policy dialog related to military AI.
Near-Earth objects and nuclear war
We are interested in the risk an explosion caused by a near-Earth object (an asteroid, comet, or meteor) colliding with Earth could be misinterpreted as a nuclear explosion and triggering a nuclear war. We have explored this topic in prior research (this and this), including through a preliminary risk model. We are interested in refining the risk analysis of this topic and further developing its implications. We are interested in the assistance of a student or early career researcher, ideally with background or at least interest in both NEOs and risk analysis, with our research on this.
Nonhumans in AI ethics
We are studying how concern for nonhumans should be incorporated in the design and use of AI systems. Our project involves fundamental topics in moral philosophy, the development and application of institutional ethics principles in governance, and the designing techniques such as inverse reinforcement learning to handle nonhumans appropriately. We are considering both “natural” nonhumans, such as nonhuman animals and ecosystems, and “artificial” nonhumans, such as AI systems themselves. We are part of a small community of ethicists and computer scientists working on this topic, and we welcome inquiries from anyone who wishes to join.
Nonhumans in global catastrophic risk
We are studying alternative conceptions of global catastrophic risk that include catastrophes affecting nonhumans. Global catastrophic risk has traditionally been defined in terms of the impact of catastrophes on humans, but arguably the impact of catastrophes on nonhumans is equally important. This is an early stage project, and we welcome inquiries from anyone with interest in the topic.
Transfer of safety techniques to AI projects
We are studying how to ensure important safety techniques developed by external groups are implemented in AI projects. Our initial work has explored both the AI governance and safety science literatures, but we have so far found very little relevant research. We welcome any suggestions, ideas, or offers of assistance.
I'm happy to answer any questions and we hope to hear from you soon!
